0023-6837/00/8010-1513$03.00/0 LABORATORY INVESTIGATION Vol. 80, No. 10, p. 1513, 2000 Copyright © 2000 by The United States and Canadian Academy of Pathology, Inc. Printed in U.S.A. DNA Methylation and the Mechanisms of CDKN2A Inactivation in Transitional Cell Carcinoma of the Urinary Bladder Andrea R. Florl, Knut H. Franke, Dieter Niederacher, Claus-Dieter Gerharz, Hans-Helge Seifert, and Wolfgang A. Schulz Urologische Klinik (ARF, KHF, H-HS, WAS), Frauenklinik (DN), and Institut für Pathologie (C-DG), Heinrich Heine Universität, Düsseldorf, Germany SUMMARY: Alterations of the CDKN2A locus on chromosome 9p21 encoding the p16INK4A cell cycle regulator and the p14ARF1 p53 activator proteins are frequently found in bladder cancer. Here, we present an analysis of 86 transitional cell carcinomas (TCC) to elucidate the mechanisms responsible for inactivation of this locus. Multiplex quantitative PCR analysis for five microsatellites around the locus showed that 34 tumors (39%) had loss of heterozygosity (LOH) generally encompassing the entire region. Of these, 17 tumors (20%) carried homozygous deletions of at least one CDKN2A exon and of flanking microsatellites, as detected by quantitative PCR. Analysis by restriction enzyme PCR and methylation-specific PCR showed that only three specimens, each with LOH across 9p21, had bona fide hypermethylation of the CDKN2A exon 1␣ CpG-island in the remaining allele. Like most other specimens, these three specimens displayed substantial genome-wide hypomethylation of DNA as reflected in the methylation status of LINE L1 sequences. The extent of DNA hypomethylation was significantly more pronounced in TCC with LOH and/or homozygous deletions at 9p21 than in those without (26% and 28%, respectively, on average, versus 11%, p Ͻ 0.0015). No association of LOH or homozygous deletions at 9p21 with tumor stage or grade was found. The data indicate that DNA hypermethylation may be rare in TCC and that deletions are the most important mechanism for inactivation of the CDKN2A locus. The predominance of allelic loss may be explained by its correlation with genome-wide DNA hypomethylation, which is thought to favor chromosomal instability and illegitimate recombination. (Lab Invest 2000, 80:1513– 1522). nactivation of the CDKN2A locus at chromosome Loss of chromosome 9 has long been recognized as I 9p21 is one of the most frequent genetic changes the most characteristic chromosomal aberration in in human cancers (Kamb et al, 1994). This unusual transitional cell carcinoma (TCC) of the urinary bladder locus encodes two distinct proteins (Quelle et al, 1995) (Knowles, 1999; Orntoft and Wolf, 1998). Many TCC involved in two pathways crucial for the regulation of display loss of one entire chromosome 9, others show cell proliferation and the control of genomic integrity. loss of regions only on the short (9p) or the long arm The p16INK4A protein, encoded by exons E1␣, E2, and (9q). Although the genes impaired by changes on 9q E3, is a cell cycle regulator that limits the activity of have not been identified, overwhelming evidence indi- cyclin-dependent kinases and inhibits entry into cates that CDKN2A is an important locus targeted by S-phase when it is accumulated. The p14ARF1 protein alterations on chromosome 9p. Homozygous dele- is encoded by E1␤, E2, and E3 in a different reading tions of this locus have been observed in several frame, is induced during cell proliferation and supports studies (Balazs et al, 1997; Baud et al, 1999; Orlow et activation of p53 by blocking the p53 inhibitor, MDM2. al, 1995, 1999; Packenham et al, 1995; Williamson et Reported mechanisms of CDKN2A inactivation in can- al, 1995) and hypermethylation of the CpG island cer cells comprise deletion, mutation, and promoter around exon E1␣ of CDKN2A has been observed in hypermethylation. The frequencies of CDKN2A inacti- cell lines and primary tumors (Akao et al, 1997; vation and the relative importance of each of these Gonzalez-Zulueta et al, 1995; Gonzalgo et al, 1998; inactivation mechanisms differ considerably between Merlo et al, 1995). Additionally, point mutations affect- different cancers (Ruas and Peters, 1998). ing both reading frames have been found, albeit with low frequency (Baud et al, 1999; Kai et al, 1995), and re-expression of p16INK4A has been shown to sup- press the growth of bladder carcinoma cells harboring Received May 25, 2000. an inactivated gene (Bender et al, 1998; Gonzalgo et This study was supported by the Deutsche Forschungsgemeinschaft al, 1998; Grim et al, 1997). (Schm782/3-1), the VERUM foundation, and the Center for Biomedical Research at the Heinrich-Heine-University. Despite ample work in this area, several issues have Address reprint requests to: Dr. W.A. Schulz, Urologische Klinik, Heinrich not been fully resolved. First, some studies in TCC and Heine Universität, Moorenstrasse 5, D-40225 Düsseldorf, Germany. Fax: other carcinomas have observed two distinct regions 49 211 81 18676; E-mail: [email protected] of loss of heterozygosity (LOH) around CDKN2A, Laboratory Investigation • October 2000 • Volume 80 • Number 10 1513 Florl et al suggesting that more than one tumor suppressor gene 1999; Jürgens et al, 1996). Although focal DNA hyper- may be located in this region (Czerniak et al, 1999). methylation contributes to the inactivation of individ- Second, although hypermethylation of the CpG-island ual genes such as CDKN2A, there is evidence that encompassing exon 1␣ is associated with transcrip- genome-wide DNA hypomethylation in tumors is as- tional repression, the frequency of hypermethylation sociated with an increased risk of illegitimate recom- reported in different studies on TCC (Akao et al, 1997; bination, gene deletion and chromosome loss (Ehrlich, Gonzalez-Zulueta et al, 1995; Gonzalgo et al, 1998; 2000; Schulz, 1998). Moreover, increased frequencies Merlo et al, 1995) has varied considerably. Third, the of chromosomal loss and gene deletions have been relationship between CDKN2A alterations and pro- observed in patients with the inherited immunodefi- gression of TCC is a matter of debate. On one hand, ciency, centromeric region instability, facial anomalies loss of CDKN2A function seems to be required for (ICF) syndrome in which hypomethylation of certain immortalization of urothelial carcinoma cells in culture repetitive sequences occurs (Jeanpierre et al, 1993; (Yeager et al, 1998). Accordingly, several studies have Xu et al, 1999), and in mice with decreased levels of suggested that such alterations are associated with methylation caused by targeted deletions of DNA tumor progression (Orlow et al, 1999). On the other methyltransferase 1 (Chen et al, 1998). Therefore, the hand, loss of chromosome 9, including the region of high prevalence of DNA hypomethylation might ex- the CDKN2A locus, has been reported in well- plain the predominance of deletions being the mech- differentiated papillary TCC (Balazs et al, 1997; Hart- anism of CDKN2A gene inactivation in TCC. mann et al, 1999; Jung et al, 1999), which rarely These issues were addressed in the present study, progress to invasiveness. based on the analysis of 86 TCC samples employing Although not all studies agree, the overall published LOH analysis by quantitative multiplex PCR, detection data on the mechanisms of CDKN2A inactivation in of homozygous deletions by quantitative PCR, DNA TCC suggest that deletions are the most common hypermethylation analysis by restriction enzyme PCR alteration in this tumor type, whereas point mutations (RE-PCR) and methylation-specific PCR (MS-PCR), or DNA hypermethylation predominate in other tumors and quantification of genome-wide hypomethylation (Ruas and Peters, 1998). Because the various modes by Southern blot analysis of LINE L1 sequence of inactivation are probably functionally equivalent, the methylation. causes for this difference must be sought in the mechanisms of mutation characteristic for each tumor type. In this regard, it may be relevant that early Results invasive TCC or their presumptive precursors such as LOH Analysis high grade papillary tumors (pTaG3) and carcinoma in situ (CIS) often show a high number of chromosomal LOH analysis was first performed using five microsat- alterations (Kallioniemi et al, 1995; Presti et al, 1991; ellite markers surrounding the CDNK2 locus and dis- Zhao et al, 1999). This chromosomal instability may be tributed across 9p21 (Fig. 1). DNA from each tumor promoted by the loss of p53 function in many TCC sample was compared with normal DNA isolated from that is associated with a high risk of progression (Esrig leukocytes or with morphologically normal corre- et al, 1994). However, TCC are also characterized by a sponding tissue. LOH analysis was performed in mul- high prevalence of genome-wide DNA hypomethyla- tiplex reactions (Fig. 2) with quantitative evaluation of tion, which is extensive in some cases (Florl et al, signal intensities as detailed in the “Materials and Figure 1. Partial map of chromosome 9p21. A, Locations of the CDKN2A and CDKN2B loci and the microsatellites analyzed. B, DNA fragments amplified by PCR are indicated. 1514 Laboratory Investigation • October 2000 • Volume 80 • Number 10 CDKN2A and DNA Methylation in Bladder Cancer Methods” section. Thirty-four tumors reproducibly E1␣ was investigated as a possible cause of gene showed LOH for at least one marker. In eight speci- inactivation. Two different methods were used, RE- mens, LOH encompassed all informative markers. PCR and MS-PCR. DNA from the bladder